MODELING OF AUTOGENOUS DEFORMATION IN CEMENTITIOUS MATERIALS, RESTRAINING EFFECT FROM AGGREGATE, AND MOISTURE WARPING IN SLABS ON GRADE by Ya Wei A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy (Civil Engineering) in The University of Michigan 2008 Doctoral Committee: Professor Will Hansen, Chair Professor Jwo Pan Associate Professor Gustavo J. Parra-Montesinos Assistant Professor Elin A. Jensen Ya Wei © 2008 All Rights Reserved This thesis is dedicated to my beloved son DongDong. ii ACKNOWLEDGEMENTS I wish to thank the many individuals who assisted me in the completion of this dissertation.
First, special thanks to my advisor, Prof. Will Hansen, for providing support, encouragements, and suggestions during this research. His knowledge and dedication have made this work possible. I also wish to thank my committee members Prof.
Jwo Pan, Prof. Parra-Montesinos, and Prof. Jensen for spending time reviewing my work and providing comments. I would also like to express my appreciation to the faculty and staff of the Department of Civil and Environmental Engineering at the University of Michigan.
The professors, technicians, and office staff have offered me endless support throughout my study here. Thank you to Phil Mohr, Yanfei Peng, and Youngjae Kang for their friendship and support. Finally, thank you to my family who are always supportive and patient. iii TABLE OF CONTENTS DEDICATION.
iii LIST OF FIGURES. vii LIST OF TABLES .2 OBJECTIVES AND SCOPE OF THIS RESEARCH. CHARACTERISTICS OF HYDRATING CEMENT PASTE: LITERATURE REVIEW .2 Development of Pore Structure.3 Shrinkage due to Chemical Reactions .2 Mechanisms of Autogenous Shrinkage .3 Experimental Techniques for the Measurement of Autogenous Deformation .4 GROUND GRANULATED BLAST-FURNACE SLAG (GGBFS) .5 STRUCTURAL RESPONSE OF A SLAB TO CEMENT HYDRATION .4 AUTOGENOUS DEFORMATION MEASUREMENT.5 SELF-INDUCED STRESS DUE TO RESTRAINED AUTOGENOUS DEFORMATION .6 CHARACTERIZATION OF HYDRATION PRODUCTS USING THERMOGRAVIMETRIC ANALYSIS (TGA) .1 Preparation of Samples .2 Chemically Bound Water .4 Calcium Silicate Hydrate (C-S-H) .7 STRENGTH AND YOUNG’S MODULUS TESTS .8 FINITE ELEMENT ANALYSIS TOOL-MLS. AUTOGENOUS DEFORMATION AND THERMOGRAVIMETRIC ANALYSIS OF CEMENT PASTE .2 AUTOGENOUS SHRINKAGE AS A RESULT OF SELF- DESICCATION DUE TO CEMENT HYDRATION .1 Effect of Water/cementitious Ratio .2 Effect of Cement Chemical Compositions .3 AUTOGENOUS SHRINKAGE OF PASTE BLENDED WITH GGBFS.3 Quantifying Pozzolanic Contribution to Autogenous Shrinkage .4 Primary Hydration Products Controlling Autogenous Shrinkage in a Blended System .5 Relating Chemically Bound Water to the Physico- Mechanical Properties of Blended System .6 Relating Autogenous Shrinkage to Chemical Shrinkage.
MODELING OF AUTOGENOUS SHRINKAGE AND AGGREGATE RESTRAINING EFFECT IN CONCRETE .2 PREDICTION OF PASTE AUTOGENOUS SHRINKAGE .2 Prediction Based on F-H Model .3 Prediction as a Function of Chemically Bound Water.3 MODELING OF CONCRETE AUTOGENOUS SHRINKAGE AND AGGREGATE RESTRAINING EFFECT .3 Application of Pickett’s Model to Concrete Autogenous Shrinkage Predictions .4 Modified Pickett’s Model .4 ASSESSMENT OF EARLY-AGE CRACKING RISK DUE TO RESTRAINED AUTOGENOUS SHRINKAGE. MOISTURE WARPING IN SLABS ON GRADE .4 FINITE ELEMENT MODELING .1 Model in MLS and Its Calibration .2 Prediction of Slab Warping .5 MECHANISM BEHIND MOISTURE WARPING .1 Theory of Depercolation of Capillary Pores .2 Understanding Moisture Warping.6 REVISED MOISTURE WARPING THEORY. PRE-SOAKED LIGHTWEIGHT FINE AGGREGATE AS ADDITIVES FOR INTERNAL CURING IN CONCRETE .2 REVIEW OF INTERNAL CURING TECHNIQUES .4 RESULTS AND DISCUSSIONS .1 Mitigation of Autogenous Shrinkage .2 Mitigation of Self-induced Stress .3 Mitigation of Moisture Warping .4 Mitigation of Drying Shrinkage .3 RECOMMENDATIONS FOR FUTURE WORK .150 vi LIST OF FIGURES Figure # Chapter 1 1.1 Flow diagram of this thesis .1 Development of hydration stages and hydration products [Locher et al.2 Morphology of the plate-like CH and fine bundles of C-S-H, and ettringite needles [Stutzman 2001] .3 Schematic illustration of the volumetric changes of sealed cement paste for w/cm=0.475 at different stages of hydration [Kovler and Zhutoysky 2006] .4 Calculated pore humidity versus degree of hydration for three different cement finenesses and w/cm of the cement paste [Koenders 1997] .5 Capillary tension as the mechanism of shrinkage .6 Volumetric measurement of autogenous shrinkage [Bjontegaard 1999] .7 Linear measurement of autogenous shrinkage [Japan 1999] .8 Illustration of shrinkage cracking in a concrete overlay .9 Slab deformation and stresses caused by a non-linear shrinkage gradient [Springenschmid 2001] .1 Linear measurement of autogenous shrinkage (a) photo of an empty rig [after Schleibinger 2007]; (b) schematic illustration of the apparatus; (c) specimens in testing .2 Top view of TSTM (a) schematic illustration of the apparatus; (b) specimens right after casting and prior to testing .3 (a) Mettler TGA (model 851 LF); (b) typical TGA and DTGA curves for a cement paste sample after 180 days of hydration .4 Temperature and stress contour in a concrete member simulated by using MLS [2003] .1 Measured autogenous shrinkage of OPC paste at three w/cm ratios (a) linear time scale; (b) logarithmic time scale .2 (a) MLS predicted reduction of pore humidity in the sealed-cured OPC paste at three w/cm ratios; (b) measured development of chemically bound water in OPC paste at three w/cm ratios .3 Relationship between measured autogenous shrinkage and (a) degree of hydration; (b) MLS predicted pore humidity of OPC paste .4 Measured internal (pore) humidity and the associated autogenous shrinkage for three concrete mixes [Jonasson et al.5 Measured autogenous shrinkage of white cement paste at three w/cm ratios (a) linear time scale; (b) logarithmic time scale.6 Measured autogenous shrinkage of paste blended with GGBFS, w/cm=0.35 (a) linear time scale; (b) logarithmic time scale.7 DTGA curves of blended paste with w/cm=0.35 at the age of (a) 1 day; (b) 90 days.8 TGA results over time for paste blended with GGBFS at w/cm=0.9 Normalization of pozzolanic contribution to autogenous shrinkage of blended cement paste, w/cm=0.10 Contribution of pozzolanic reactions to autogenous shrinkage as a function of (a) time; (b) degree of hydration of OPC, w/cm=0.11 (a) PHP loss as a function of degree of hydration of OPC; (b) CH loss as a function of degree of hydration of OPC in a blended paste, w/cm= 0.12 Compressive strength development of concrete blended with GGBFS, w/cm=0.13 Relationship between autogenous shrinkage and PHPloss for blended systems, w/cm=0.14 Relationship between autogenous shrinkage and CHloss for blended systems, w/cm=0.15 TGA results of various phases in blended systems with w/cm=0.16 Relationship between autogenous shrinkage and Wn loss for blended systems with w/cm= 0.17 Relationship between autogenous shrinkage and Wn loss for systems with different w/cm .18 Relationship between compressive strength and Wn loss for concrete blended with GGBFS, w/cm=0.19 Chemical and autogenous shrinkage of cement paste with w/cm=0.4 [Sellevold et al.20 Relationship between measured autogenous shrinkage and predicted chemical shrinkage for OPC paste .21 Relationship between measured autogenous shrinkage and predicted chemical shrinkage for blended systems, w/cm=0.22 The reduction of pore humidity with chemical shrinkage for OPC paste.1 Measured autogenous shrinkage and the curve fit using F-H model for OPC paste .2 Measured autogenous shrinkage and the curve fit using F-H model for white cement paste .3 Measured autogenous shrinkage and the curve fit using F-H model for blended cement paste .4 Parameters in F-H model vs. w/cm for predicting autogenous shrinkage of OPC paste.5 Parameters in F-H model vs.
w/cm for predicting autogenous shrinkage of white cement paste .6 Parameters in F-H model vs. GGBFS content for predicting autogenous shrinkage of blended paste, w/cm=0.7 Autogenous shrinkage as a function of chemically bound water for blended cement paste, w/cm=0.8 Chemically bound water as a function of time for blended cement paste, w/cm=0.9 Parameters in F-H model vs. GGBFS content for predicting chemically bound water of blended paste, w/cm=0.10 Measured and predicted autogenous shrinkage using chemically bound water for blended cement paste, w/cm=0.11 Parameters (obtained through chemically bound water) in F-H model vs. GGBFS content for predicting autogenous shrinkage of blended paste, w/cm=0.12 Measured autogenous shrinkage of cement paste and concrete with different aggregate contents (a) w/cm=0.13 Normalized autogenous shrinkage of cement paste and concrete with different aggregate contents (a) w/cm=0.14 Schematic illustration of the small spherical aggregate particle within a concrete sphere [after Hansen and Nielsen 1965].15 Measured Young’s modulus for portland cement concrete, aggregate content =57%.16 Measured and predicted autogenous shrinkage using Pickett’s model for concrete with w/cm=0.17 Measured and predicted autogenous shrinkage using Pickett’s model for concrete with w/cm=0.18 Measured and predicted autogenous shrinkage using Pickett’s mode for concrete with w/cm=0.19 Schematic illustration of the microcracking generated on the interface of the nonshrinking spherical aggregate and the shrinking body .20 Relationship between aggregate content and autogenous shrinkage of concrete at different ages for (a) w/cm=0.21 Predicted n values in modified Pickett’s model .22 Parameters needed for prediction of n .23 Comparison of Pickett’s model and modified Pickett’s model for autogenous shrinkage predictions (a) w/cm=0.24 Development of temperature and stresses in a restrained concrete specimen .25 Schematic illustration of the generation of tensile stress from restrained autogenous shrinkage deformation .26 Measured free autogenous deformation for concrete with φ A =57% (a) w/cm=0.27 Measured stress from restrained autogenous deformation for concrete with φ A =57% (a) w/cm=0.28 Relationship between free shrinkage and self-induced stress, w/cm=0.29 Illustration of the stress status in concrete when subjected to (a) tension; (b) compression .1 Development of pre-mature top-down transverse cracking in two JPCP projects .2 Finite element-based (EverFE) rendering of uplift for a JPCP project .3 Surface elevation profiles for JPCPs in southeastern Michigan with matching TELTD values listed .4 Laboratory moisture warping test .5 Moisture conditions simulated in beam warping test .6 Measured beam warping results, w/cm=0.7 Finite element modeling of beam warping test .8 FE calibration results for beam warping test, w/cm=0.9 Slab model in MLS .10 Predicted slab warping under three moisture conditions, w/cm=0.11 (a) percolation and (b) depercolation of capillary pores in hydrating paste .12 Measured permeability changes over time by different authors .13 Predicted (a) uniform moisture gradient from sealed curing; (b) non-uniform moisture gradient after 28 days of exposure to drying at slab top following 28 days of sealed curing; (c) combined moisture gradient after 28 days of exposure to drying at top and moisture wetting at slab bottom following 28 days of sealed curing, w/cm=0.14 Sketch of slab deformations under different moisture gradients .1 Scanning electron image of mortar containing lightweight sand [Lam 2005].2 Measured development of autogenous deformation (internal curing vs.
no internal curing) for concretes with φ A =57% and (a) w/cm=0.3 Measured self-induced stress (internal curing vs. no internal curing) for concretes with φ A =57% and (a) w/cm=0.4 Measured development of beam warping (internal curing vs. no internal curing) for w/cm=0.5 Measured development of drying shrinkage (internal curing vs. no internal curing) for w/cm=0.144 xii LIST OF TABLES Table # Chapter 3 3.1 Physical properties and chemical compositions of cementitious materials .2 Mix proportioning for paste ( φ A =0%) and concrete .3 Mix proportioning of cement paste blended with GGBFS, w/cm=0.4 Mix proportioning of concrete containing pre-soaked lightweight fine aggregate (LWFA) at different replacement levels, w/cm=0.1 Hygral parameters used in MLS modeling .2 Approximate age for the capillary pores to become discontinuous [Powers et al.128 xiii CHAPTER 1 INTRODUCTION 1.1 PROBLEM STATEMENT Concrete infrastructures undergo complex chemical and physical changes from cement hydration and exposure to the environments.
These changes will affect the desired service life or durability of the hydraulic-cement concrete. In recent years, the durability problem of infrastructures has become a major concern since many of them are in serious need of repair, retrofitting, or replacement. There is increasing pressure to ensure that new constructions remain serviceable condition for long periods with only minimum maintenance. There are many factors affecting the durability of the cementitious materials and the importance of these factors is varying with circumstances.
Shrinkage is one of the major causes for cracking in bridge decks, pavements, indoor floors and other structures. Concrete develops volumetric changes due to the thermal and moisture related deformations which can be detrimental when substantial stresses occur in restrained structural elements, particularly at early ages while the concrete has a low tensile strength.